Margrit Betke

Donghyun Kim, Kaihong Wang, Kate Saenko et al.

While pose estimation is an important computer vision task, it requires expensive annotation and suffers from domain shift. In this paper, we investigate the problem of domain adaptive 2D pose estimation that transfers knowledge learned on a synthetic source domain to a target domain without supervision. While several domain adaptive pose estimation models have been proposed recently, they are not generic but only focus on either human pose or animal pose estimation, and thus their effectiveness is somewhat limited to specific scenarios. In this work, we propose a unified framework that generalizes well on various domain adaptive pose estimation problems. We propose to align representations using both input-level and output-level cues (pixels and pose labels, respectively), which facilitates the knowledge transfer from the source domain to the unlabeled target domain. Our experiments show that our method achieves state-of-the-art performance under various domain shifts. Our method outperforms existing baselines on human pose estimation by up to 4.5 percent points (pp), hand pose estimation by up to 7.4 pp, and animal pose estimation by up to 4.8 pp for dogs and 3.3 pp for sheep. These results suggest that our method is able to mitigate domain shift on diverse tasks and even unseen domains and objects (e.g., trained on horse and tested on dog). Our code will be publicly available at: https://github.com/VisionLearningGroup/UDA_PoseEstimation.

Yi Zheng, Rushin H. Gindra, Emily J. Green et al.

Deep learning is a powerful tool for whole slide image (WSI) analysis. Typically, when performing supervised deep learning, a WSI is divided into small patches, trained and the outcomes are aggregated to estimate disease grade. However, patch-based methods introduce label noise during training by assuming that each patch is independent with the same label as the WSI and neglect overall WSI-level information that is significant in disease grading. Here we present a Graph-Transformer (GT) that fuses a graph-based representation of an WSI and a vision transformer for processing pathology images, called GTP, to predict disease grade. We selected $4,818$ WSIs from the Clinical Proteomic Tumor Analysis Consortium (CPTAC), the National Lung Screening Trial (NLST), and The Cancer Genome Atlas (TCGA), and used GTP to distinguish adenocarcinoma (LUAD) and squamous cell carcinoma (LSCC) from adjacent non-cancerous tissue (normal). First, using NLST data, we developed a contrastive learning framework to generate a feature extractor. This allowed us to compute feature vectors of individual WSI patches, which were used to represent the nodes of the graph followed by construction of the GTP framework. Our model trained on the CPTAC data achieved consistently high performance on three-label classification (normal versus LUAD versus LSCC: mean accuracy$= 91.2$ $\pm$ $2.5\%$) based on five-fold cross-validation, and mean accuracy $= 82.3$ $\pm$ $1.0\%$ on external test data (TCGA). We also introduced a graph-based saliency mapping technique, called GraphCAM, that can identify regions that are highly associated with the class label. Our findings demonstrate GTP as an interpretable and effective deep learning framework for WSI-level classification.

Zhaodan Kong, Kayhan Özcimder, Nathan Fuller et al.

The movements of birds, bats, and other flying species are governed by complex sensorimotor systems that allow the animals to react to stationary environmental features as well as to wind disturbances, other animals in nearby airspace, and a wide variety of unexpected challenges. The paper and talk will describe research that analyzes the three-dimensional trajectories of bats flying in a habitat in Texas. The trajectories are computed with stereoscopic methods using data from synchronous thermal videos that were recorded with high temporal and spatial resolution from three viewpoints. Following our previously reported work, we examine the possibility that bat trajectories in this habitat are governed by optical flow sensing that interpolates periodic distance measurements from echolocation. Using an idealized geometry of bat eyes, we introduce the concept of time-to-transit, and recall some research that suggests that this quantity is computed by the animals' visual cortex. Several steering control laws based on time-to-transit are proposed for an idealized flight model, and it is shown that these can be used to replicate the observed flight of what we identify as typical bats. Although the vision-based motion control laws we propose and the protocols for switching between them are quite simple, some of the trajectories that have been synthesized are qualitatively bat-like. Examination of the control protocols that generate these trajectories suggests that bat motions are governed both by their reactions to a subset of key feature points as well by their memories of where these feature points are located.

Kaihong Wang, Donghyun Kim, Rogerio Feris et al.

While transformers have greatly boosted performance in semantic segmentation, domain adaptive transformers are not yet well explored. We identify that the domain gap can cause discrepancies in self-attention. Due to this gap, the transformer attends to spurious regions or pixels, which deteriorates accuracy on the target domain. We propose to perform adaptation on attention maps with cross-domain attention layers that share features between the source and the target domains. Specifically, we impose consistency between predictions from cross-domain attention and self-attention modules to encourage similar distribution in the attention and output of the model across domains, i.e., attention-level and output-level alignment. We also enforce consistency in attention maps between different augmented views to further strengthen the attention-based alignment. Combining these two components, our method mitigates the discrepancy in attention maps across domains and further boosts the performance of the transformer under unsupervised domain adaptation settings. Our model outperforms the existing state-of-the-art baseline model on three widely used benchmarks, including GTAV-to-Cityscapes by 1.3 percent point (pp), Synthia-to-Cityscapes by 0.6 pp, and Cityscapes-to-ACDC by 1.1 pp, on average. Additionally, we verify the effectiveness and generalizability of our method through extensive experiments. Our code will be publicly available.

Ge Gao, Jongin Kim, Sejin Paik et al.

Predicting emotions elicited by news headlines can be challenging as the task is largely influenced by the varying nature of people's interpretations and backgrounds. Previous works have explored classifying discrete emotions directly from news headlines. We provide a different approach to tackling this problem by utilizing people's explanations of their emotion, written in free-text, on how they feel after reading a news headline. Using the dataset BU-NEmo+ (Gao et al., 2022), we found that for emotion classification, the free-text explanations have a strong correlation with the dominant emotion elicited by the headlines. The free-text explanations also contain more sentimental context than the news headlines alone and can serve as a better input to emotion classification models. Therefore, in this work we explored generating emotion explanations from headlines by training a sequence-to-sequence transformer model and by using pretrained large language model, ChatGPT (GPT-4). We then used the generated emotion explanations for emotion classification. In addition, we also experimented with training the pretrained T5 model for the intermediate task of explanation generation before fine-tuning it for emotion classification. Using McNemar's significance test, methods that incorporate GPT-generated free-text emotion explanations demonstrated significant improvement (P-value < 0.05) in emotion classification from headlines, compared to methods that only use headlines. This underscores the value of using intermediate free-text explanations for emotion prediction tasks with headlines.

Zijian Chen, Maria Varkanitsa, Prakash Ishwar et al.

We propose a lesion-aware graph neural network (LEGNet) to predict language ability from resting-state fMRI (rs-fMRI) connectivity in patients with post-stroke aphasia. Our model integrates three components: an edge-based learning module that encodes functional connectivity between brain regions, a lesion encoding module, and a subgraph learning module that leverages functional similarities for prediction. We use synthetic data derived from the Human Connectome Project (HCP) for hyperparameter tuning and model pretraining. We then evaluate the performance using repeated 10-fold cross-validation on an in-house neuroimaging dataset of post-stroke aphasia. Our results demonstrate that LEGNet outperforms baseline deep learning methods in predicting language ability. LEGNet also exhibits superior generalization ability when tested on a second in-house dataset that was acquired under a slightly different neuroimaging protocol. Taken together, the results of this study highlight the potential of LEGNet in effectively learning the relationships between rs-fMRI connectivity and language ability in a patient cohort with brain lesions for improved post-stroke aphasia evaluation.

Mahir Patel, Lucas Carstensen, Yiwen Gu et al.

A major bottleneck of interdisciplinary computer vision (CV) research is the lack of a framework that eases the reuse and abstraction of state-of-the-art CV models by CV and non-CV researchers alike. We present here BU-CVKit, a computer vision framework that allows the creation of research pipelines with chainable Processors. The community can create plugins of their work for the framework, hence improving the re-usability, accessibility, and exposure of their work with minimal overhead. Furthermore, we provide MuSeqPose Kit, a user interface for the pose estimation package of BU-CVKit, which automatically scans for installed plugins and programmatically generates an interface for them based on the metadata provided by the user. It also provides software support for standard pose estimation features such as annotations, 3D reconstruction, reprojection, and camera calibration. Finally, we show examples of behavioral neuroscience pipelines created through the sample plugins created for our framework.

Alyssa Smith, David Assefa Tofu, Mona Jalal et al.

When journalists cover a news story, they can cover the story from multiple angles or perspectives. A news article written about COVID-19 for example, might focus on personal preventative actions such as mask-wearing, while another might focus on COVID-19's impact on the economy. These perspectives are called "frames," which when used may influence public perception and opinion of the issue. We introduce a Web-based system for analyzing and classifying frames in text documents. Our goal is to make effective tools for automatic frame discovery and labeling based on topic modeling and deep learning widely accessible to researchers from a diverse array of disciplines. To this end, we provide both state-of-the-art pre-trained frame classification models on various issues as well as a user-friendly pipeline for training novel classification models on user-provided corpora. Researchers can submit their documents and obtain frames of the documents. The degree of user involvement is flexible: they can run models that have been pre-trained on select issues; submit labeled documents and train a new model for frame classification; or submit unlabeled documents and obtain potential frames of the documents. The code making up our system is also open-sourced and well-documented, making the system transparent and expandable. The system is available on-line at http://www.openframing.org and via our GitHub page https://github.com/davidatbu/openFraming .

Mona Jalal, Kate K. Mays, Lei Guo et al.

We report results of a comparison of the accuracy of crowdworkers and seven Natural Language Processing (NLP) toolkits in solving two important NLP tasks, named-entity recognition (NER) and entity-level sentiment (ELS) analysis. We here focus on a challenging dataset, 1,000 political tweets that were collected during the U.S. presidential primary election in February 2016. Each tweet refers to at least one of four presidential candidates, i.e., four named entities. The groundtruth, established by experts in political communication, has entity-level sentiment information for each candidate mentioned in the tweet. We tested several commercial and open-source tools. Our experiments show that, for our dataset of political tweets, the most accurate NER system, Google Cloud NL, performed almost on par with crowdworkers, but the most accurate ELS analysis system, TensiStrength, did not match the accuracy of crowdworkers by a large margin of more than 30 percent points.

Hao Yu, Rupayan Mallick, Margrit Betke et al.

Cartoon avatars have been widely used in various applications, including social media, online tutoring, and gaming. However, existing cartoon avatar datasets and generation methods struggle to present highly expressive avatars with fine-grained facial expressions and are often inspired from real-world identities, raising privacy concerns. To address these challenges, we propose a novel framework, GenEAva, for generating high-quality cartoon avatars with fine-grained facial expressions. Our approach fine-tunes a state-of-the-art text-to-image diffusion model to synthesize highly detailed and expressive facial expressions. We then incorporate a stylization model that transforms these realistic faces into cartoon avatars while preserving both identity and expression. Leveraging this framework, we introduce the first expressive cartoon avatar dataset, GenEAva 1.0, specifically designed to capture 135 fine-grained facial expressions, featuring 13,230 expressive cartoon avatars with a balanced distribution across genders, racial groups, and age ranges. We demonstrate that our fine-tuned model generates more expressive faces than the state-of-the-art text-to-image diffusion model SDXL. We also verify that the cartoon avatars generated by our framework do not include memorized identities from fine-tuning data. The proposed framework and dataset provide a diverse and expressive benchmark for future research in cartoon avatar generation.

Wenda Qin, Andrea Burns, Bryan A. Plummer et al.

Large models achieve strong performance on Vision-and-Language Navigation (VLN) tasks, but are costly to run in resource-limited environments. Token pruning offers appealing tradeoffs for efficiency with minimal performance loss by reducing model input size, but prior work overlooks VLN-specific challenges. For example, information loss from pruning can effectively increase computational cost due to longer walks. Thus, the inability to identify uninformative tokens undermines the supposed efficiency gains from pruning. To address this, we propose Navigation-Aware Pruning (NAP), which uses navigation-specific traits to simplify the pruning process by pre-filtering tokens into foreground and background. For example, image views are filtered based on whether the agent can navigate in that direction. We also extract navigation-relevant instructions using a Large Language Model. After filtering, we focus pruning on background tokens, minimizing information loss. To further help avoid increases in navigation length, we discourage backtracking by removing low-importance navigation nodes. Experiments on standard VLN benchmarks show NAP significantly outperforms prior work, preserving higher success rates while saving more than 50% FLOPS.

Hao Yu, Rupayan Mallick, Margrit Betke et al.

Different forms of customized 2D avatars are widely used in gaming applications, virtual communication, education, and content creation. However, existing approaches often fail to capture fine-grained facial expressions and struggle to preserve identity across different expressions. We propose GEN-AFFECT, a novel framework for personalized avatar generation that generates expressive and identity-consistent avatars with a diverse set of facial expressions. Our framework proposes conditioning a multimodal diffusion transformer on an extracted identity-expression representation. This enables identity preservation and representation of a wide range of facial expressions. GEN-AFFECT additionally employs consistent attention at inference for information sharing across the set of generated expressions, enabling the generation process to maintain identity consistency over the array of generated fine-grained expressions. GEN-AFFECT demonstrates superior performance compared to previous state-of-the-art methods on the basis of the accuracy of the generated expressions, the preservation of the identity and the consistency of the target identity across an array of fine-grained facial expressions.

Kaihong Wang, Donghyun Kim, Margrit Betke

Continual learning for vision-language models has achieved remarkable performance through synthetic replay, where samples are generated using Stable Diffusion to regularize during finetuning and retain knowledge. However, real-world downstream applications often exhibit domain-specific nuances and fine-grained semantics not captured by generators, causing synthetic-replay methods to produce misaligned samples that misguide finetuning and undermine retention of prior knowledge. In this work, we propose a LoRA-enhanced synthetic-replay framework that injects task-specific low-rank adapters into a frozen Stable Diffusion model, efficiently capturing each new task's unique visual and semantic patterns. Specifically, we introduce a two-stage, confidence-based sample selection: we first rank real task data by post-finetuning VLM confidence to focus LoRA finetuning on the most representative examples, then generate synthetic samples and again select them by confidence for distillation. Our approach integrates seamlessly with existing replay pipelines-simply swap in the adapted generator to boost replay fidelity. Extensive experiments on the Multi-domain Task Incremental Learning (MTIL) benchmark show that our method outperforms previous synthetic-replay techniques, achieving an optimal balance among plasticity, stability, and zero-shot capability. These results demonstrate the effectiveness of generator adaptation via LoRA for robust continual learning in VLMs.

Sarah Bonna, Yu-Cheng Huang, Ekaterina Novozhilova et al.

Ethical intervention prompting has emerged as a tool to counter demographic biases of text-to-image generative AI models. Existing solutions either require to retrain the model or struggle to generate images that reflect desired distributions on gender and race. We propose an inference-time process called DebiasPI for Debiasing-by-Prompt-Iteration that provides prompt intervention by enabling the user to control the distributions of individuals' demographic attributes in image generation. DebiasPI keeps track of which attributes have been generated either by probing the internal state of the model or by using external attribute classifiers. Its control loop guides the text-to-image model to select not yet sufficiently represented attributes, With DebiasPI, we were able to create images with equal representations of race and gender that visualize challenging concepts of news headlines. We also experimented with the attributes age, body type, profession, and skin tone, and measured how attributes change when our intervention prompt targets the distribution of an unrelated attribute type. We found, for example, if the text-to-image model is asked to balance racial representation, gender representation improves but the skin tone becomes less diverse. Attempts to cover a wide range of skin colors with various intervention prompts showed that the model struggles to generate the palest skin tones. We conducted various ablation studies, in which we removed DebiasPI's attribute control, that reveal the model's propensity to generate young, male characters. It sometimes visualized career success by generating two-panel images with a pre-success dark-skinned person becoming light-skinned with success, or switching gender from pre-success female to post-success male, thus further motivating ethical intervention prompting with DebiasPI.

Yiwen Gu, Mahir Patel, Margrit Betke

In this paper, we present a contrastive learning based framework, ExeChecker, for the interpretation of rehabilitation exercises. Our work builds upon state-of-the-art advances in the area of human pose estimation, graph-attention neural networks, and transformer interpretablity. The downstream task is to assist rehabilitation by providing informative feedback to users while they are performing prescribed exercises. We utilize a contrastive learning strategy during training. Given a tuple of correctly and incorrectly executed exercises, our model is able to identify and highlight those joints that are involved in an incorrect movement and thus require the user's attention. We collected an in-house dataset, ExeCheck, with paired recordings of both correct and incorrect execution of exercises. In our experiments, we tested our method on this dataset as well as the UI-PRMD dataset and found ExeCheck outperformed the baseline method using pairwise sequence alignment in identifying joints of physical relevance in rehabilitation exercises.

Isidora Chara Tourni, Lei Guo, Hengchang Hu et al.

News media structure their reporting of events or issues using certain perspectives. When describing an incident involving gun violence, for example, some journalists may focus on mental health or gun regulation, while others may emphasize the discussion of gun rights. Such perspectives are called \say{frames} in communication research. We study, for the first time, the value of combining lead images and their contextual information with text to identify the frame of a given news article. We observe that using multiple modes of information(article- and image-derived features) improves prediction of news frames over any single mode of information when the images are relevant to the frames of the headlines. We also observe that frame image relevance is related to the ease of conveying frames via images, which we call frame concreteness. Additionally, we release the first multimodal news framing dataset related to gun violence in the U.S., curated and annotated by communication researchers. The dataset will allow researchers to further examine the use of multiple information modalities for studying media framing.

Kaihong Wang, Chenhongyi Yang, Margrit Betke

Unsupervised domain adaptation for semantic segmentation has been intensively studied due to the low cost of the pixel-level annotation for synthetic data. The most common approaches try to generate images or features mimicking the distribution in the target domain while preserving the semantic contents in the source domain so that a model can be trained with annotations from the latter. However, such methods highly rely on an image translator or feature extractor trained in an elaborated mechanism including adversarial training, which brings in extra complexity and instability in the adaptation process. Furthermore, these methods mainly focus on taking advantage of the labeled source dataset, leaving the unlabeled target dataset not fully utilized. In this paper, we propose a bidirectional style-induced domain adaptation method, called BiSIDA, that employs consistency regularization to efficiently exploit information from the unlabeled target domain dataset, requiring only a simple neural style transfer model. BiSIDA aligns domains by not only transferring source images into the style of target images but also transferring target images into the style of source images to perform high-dimensional perturbation on the unlabeled target images, which is crucial to the success in applying consistency regularization in segmentation tasks. Extensive experiments show that our BiSIDA achieves new state-of-the-art on two commonly-used synthetic-to-real domain adaptation benchmarks: GTA5-to-CityScapes and SYNTHIA-to-CityScapes.

Mona Jalal, Josef Spjut, Ben Boudaoud et al.

We present a new, publicly-available image dataset generated by the NVIDIA Deep Learning Data Synthesizer intended for use in object detection, pose estimation, and tracking applications. This dataset contains 144k stereo image pairs that synthetically combine 18 camera viewpoints of three photorealistic virtual environments with up to 10 objects (chosen randomly from the 21 object models of the YCB dataset [1]) and flying distractors. Object and camera pose, scene lighting, and quantity of objects and distractors were randomized. Each provided view includes RGB, depth, segmentation, and surface normal images, all pixel level. We describe our approach for domain randomization and provide insight into the decisions that produced the dataset.

Nataniel Ruiz, Hao Yu, Danielle A. Allessio et al.

In this work, we propose a video-based transfer learning approach for predicting problem outcomes of students working with an intelligent tutoring system (ITS). By analyzing a student's face and gestures, our method predicts the outcome of a student answering a problem in an ITS from a video feed. Our work is motivated by the reasoning that the ability to predict such outcomes enables tutoring systems to adjust interventions, such as hints and encouragement, and to ultimately yield improved student learning. We collected a large labeled dataset of student interactions with an intelligent online math tutor consisting of 68 sessions, where 54 individual students solved 2,749 problems. The dataset is public and available at https://www.cs.bu.edu/faculty/betke/research/learning/ . Working with this dataset, our transfer-learning challenge was to design a representation in the source domain of pictures obtained "in the wild" for the task of facial expression analysis, and transferring this learned representation to the task of human behavior prediction in the domain of webcam videos of students in a classroom environment. We developed a novel facial affect representation and a user-personalized training scheme that unlocks the potential of this representation. We designed several variants of a recurrent neural network that models the temporal structure of video sequences of students solving math problems. Our final model, named ATL-BP for Affect Transfer Learning for Behavior Prediction, achieves a relative increase in mean F-score of 50% over the state-of-the-art method on this new dataset.

Chenhongyi Yang, Vitaly Ablavsky, Kaihong Wang et al.

While visual object detection with deep learning has received much attention in the past decade, cases when heavy intra-class occlusions occur have not been studied thoroughly. In this work, we propose a Non-Maximum-Suppression (NMS) algorithm that dramatically improves the detection recall while maintaining high precision in scenes with heavy occlusions. Our NMS algorithm is derived from a novel embedding mechanism, in which the semantic and geometric features of the detected boxes are jointly exploited. The embedding makes it possible to determine whether two heavily-overlapping boxes belong to the same object in the physical world. Our approach is particularly useful for car detection and pedestrian detection in urban scenes where occlusions often happen. We show the effectiveness of our approach by creating a model called SG-Det (short for Semantics and Geometry Detection) and testing SG-Det on two widely-adopted datasets, KITTI and CityPersons for which it achieves state-of-the-art performance.

Qitong Wang, Yi Zheng, Margrit Betke

Understanding the meaning of text in images of natural scenes like highway signs or store front emblems is particularly challenging if the text is foreshortened in the image or the letters are artistically distorted. We introduce a pipeline-based text spotting framework that can both detect and recognize text in various fonts, shapes, and orientations in natural scene images with complicated backgrounds. The main contribution of our work is the text detection component, which we call UHT, short for UNet, Heatmap, and Textfill. UHT uses a UNet to compute heatmaps for candidate text regions and a textfill algorithm to produce tight polygonal boundaries around each word in the candidate text. Our method trains the UNet with groundtruth heatmaps that we obtain from text bounding polygons provided by groundtruth annotations. Our text spotting framework, called UHTA, combines UHT with the state-of-the-art text recognition system ASTER. Experiments on four challenging and public scene-text-detection datasets (Total-Text, SCUT-CTW1500, MSRA-TD500, and COCO-Text) show the effectiveness and generalization ability of UHT in detecting not only multilingual (potentially rotated) straight but also curved text in scripts of multiple languages. Our experimental results of UHTA on the Total-Text dataset show that UHTA outperforms four state-of-the-art text spotting frameworks by at least 9.1 percent points in the F-measure, which suggests that UHTA may be used as a complete text detection and recognition system in real applications.

Kaihong Wang, Mona Jalal, Sankara Jefferson et al.

Monitoring population-level changes in diet could be useful for education and for implementing interventions to improve health. Research has shown that data from social media sources can be used for monitoring dietary behavior. We propose a scrape-by-location methodology to create food image datasets from Instagram posts. We used it to collect 3.56 million images over a period of 20 days in March 2019. We also propose a scrape-by-keywords methodology and used it to scrape ~30,000 images and their captions of 38 Kenyan food types. We publish two datasets of 104,000 and 8,174 image/caption pairs, respectively. With the first dataset, Kenya104K, we train a Kenyan Food Classifier, called KenyanFC, to distinguish Kenyan food from non-food images posted in Kenya. We used the second dataset, KenyanFood13, to train a classifier KenyanFTR, short for Kenyan Food Type Recognizer, to recognize 13 popular food types in Kenya. The KenyanFTR is a multimodal deep neural network that can identify 13 types of Kenyan foods using both images and their corresponding captions. Experiments show that the average top-1 accuracy of KenyanFC is 99% over 10,400 tested Instagram images and of KenyanFTR is 81% over 8,174 tested data points. Ablation studies show that three of the 13 food types are particularly difficult to categorize based on image content only and that adding analysis of captions to the image analysis yields a classifier that is 9 percent points more accurate than a classifier that relies only on images. Our food trend analysis revealed that cakes and roasted meats were the most popular foods in photographs on Instagram in Kenya in March 2019.

Yi Zheng, Qitong Wang, Margrit Betke

State-of-the-art text spotting systems typically aim to detect isolated words or word-by-word text in images of natural scenes and ignore the semantic coherence within a region of text. However, when interpreted together, seemingly isolated words may be easier to recognize. On this basis, we propose a novel "semantic-based text recognition" (STR) deep learning model that reads text in images with the help of understanding context. STR consists of several modules. We introduce the Text Grouping and Arranging (TGA) algorithm to connect and order isolated text regions. A text-recognition network interprets isolated words. Benefiting from semantic information, a sequenceto-sequence network model efficiently corrects inaccurate and uncertain phrases produced earlier in the STR pipeline. We present experiments on two new distinct datasets that contain scanned catalog images of interior designs and photographs of protesters with hand-written signs, respectively. Our results show that our STR model outperforms a baseline method that uses state-of-the-art single-wordrecognition techniques on both datasets. STR yields a high accuracy rate of 90% on the catalog images and 71% on the more difficult protest images, suggesting its generality in recognizing text.

Danna Gurari, Yinan Zhao, Suyog Dutt Jain et al.

Foreground object segmentation is a critical step for many image analysis tasks. While automated methods can produce high-quality results, their failures disappoint users in need of practical solutions. We propose a resource allocation framework for predicting how best to allocate a fixed budget of human annotation effort in order to collect higher quality segmentations for a given batch of images and automated methods. The framework is based on a prediction module that estimates the quality of given algorithm-drawn segmentations. We demonstrate the value of the framework for two novel tasks related to predicting how to distribute annotation efforts between algorithms and humans. Specifically, we develop two systems that automatically decide, for a batch of images, when to recruit humans versus computers to create 1) coarse segmentations required to initialize segmentation tools and 2) final, fine-grained segmentations. Experiments demonstrate the advantage of relying on a mix of human and computer efforts over relying on either resource alone for segmenting objects in images coming from three diverse modalities (visible, phase contrast microscopy, and fluorescence microscopy).

Mehrnoosh Sameki, Sha Lai, Kate K. Mays et al.

Due to concerns about human error in crowdsourcing, it is standard practice to collect labels for the same data point from multiple internet workers. We here show that the resulting budget can be used more effectively with a flexible worker assignment strategy that asks fewer workers to analyze easy-to-label data and more workers to analyze data that requires extra scrutiny. Our main contribution is to show how the allocations of the number of workers to a task can be computed optimally based on task features alone, without using worker profiles. Our target tasks are delineating cells in microscopy images and analyzing the sentiment toward the 2016 U.S. presidential candidates in tweets. We first propose an algorithm that computes budget-optimized crowd worker allocation (BUOCA). We next train a machine learning system (BUOCA-ML) that predicts an optimal number of crowd workers needed to maximize the accuracy of the labeling. We show that the computed allocation can yield large savings in the crowdsourcing budget (up to 49 percent points) while maintaining labeling accuracy. Finally, we envisage a human-machine system for performing budget-optimized data analysis at a scale beyond the feasibility of crowdsourcing.

Elham Saraee, Mona Jalal, Margrit Betke

Visual complexity identifies the level of intricacy and details in an image or the level of difficulty to describe the image. It is an important concept in a variety of areas such as cognitive psychology, computer vision and visualization, and advertisement. Yet, efforts to create large, downloadable image datasets with diverse content and unbiased groundtruthing are lacking. In this work, we introduce Savoias, a visual complexity dataset that compromises of more than 1,400 images from seven image categories relevant to the above research areas, namely Scenes, Advertisements, Visualization and infographics, Objects, Interior design, Art, and Suprematism. The images in each category portray diverse characteristics including various low-level and high-level features, objects, backgrounds, textures and patterns, text, and graphics. The ground truth for Savoias is obtained by crowdsourcing more than 37,000 pairwise comparisons of images using the forced-choice methodology and with more than 1,600 contributors. The resulting relative scores are then converted to absolute visual complexity scores using the Bradley-Terry method and matrix completion. When applying five state-of-the-art algorithms to analyze the visual complexity of the images in the Savoias dataset, we found that the scores obtained from these baseline tools only correlate well with crowdsourced labels for abstract patterns in the Suprematism category (Pearson correlation r=0.84). For the other categories, in particular, the objects and advertisement categories, low correlation coefficients were revealed (r=0.3 and 0.56, respectively). These findings suggest that (1) state-of-the-art approaches are mostly insufficient and (2) Savoias enables category-specific method development, which is likely to improve the impact of visual complexity analysis on specific application areas, including computer vision.

Danna Gurari, Kun He, Bo Xiong et al.

We propose the ambiguity problem for the foreground object segmentation task and motivate the importance of estimating and accounting for this ambiguity when designing vision systems. Specifically, we distinguish between images which lead multiple annotators to segment different foreground objects (ambiguous) versus minor inter-annotator differences of the same object. Taking images from eight widely used datasets, we crowdsource labeling the images as "ambiguous" or "not ambiguous" to segment in order to construct a new dataset we call STATIC. Using STATIC, we develop a system that automatically predicts which images are ambiguous. Experiments demonstrate the advantage of our prediction system over existing saliency-based methods on images from vision benchmarks and images taken by blind people who are trying to recognize objects in their environment. Finally, we introduce a crowdsourcing system to achieve cost savings for collecting the diversity of all valid "ground truth" foreground object segmentations by collecting extra segmentations only when ambiguity is expected. Experiments show our system eliminates up to 47% of human effort compared to existing crowdsourcing methods with no loss in capturing the diversity of ground truths.

Mikhail Breslav, Tyson L. Hedrick, Stan Sclaroff et al.

Image and video analysis is often a crucial step in the study of animal behavior and kinematics. Often these analyses require that the position of one or more animal landmarks are annotated (marked) in numerous images. The process of annotating landmarks can require a significant amount of time and tedious labor, which motivates the need for algorithms that can automatically annotate landmarks. In the community of scientists that use image and video analysis to study the 3D flight of animals, there has been a trend of developing more automated approaches for annotating landmarks, yet they fall short of being generally applicable. Inspired by the success of Deep Neural Networks (DNNs) on many problems in the field of computer vision, we investigate how suitable DNNs are for accurate and automatic annotation of landmarks in video datasets representative of those collected by scientists studying animals. Our work shows, through extensive experimentation on videos of hawkmoths, that DNNs are suitable for automatic and accurate landmark localization. In particular, we show that one of our proposed DNNs is more accurate than the current best algorithm for automatic localization of landmarks on hawkmoth videos. Moreover, we demonstrate how these annotations can be used to quantitatively analyze the 3D flight of a hawkmoth. To facilitate the use of DNNs by scientists from many different fields, we provide a self contained explanation of what DNNs are, how they work, and how to apply them to other datasets using the freely available library Caffe and supplemental code that we provide.

Mehrnoosh Sameki, Mattia Gentil, Kate K. Mays et al.

Opinions about the 2016 U.S. Presidential Candidates have been expressed in millions of tweets that are challenging to analyze automatically. Crowdsourcing the analysis of political tweets effectively is also difficult, due to large inter-rater disagreements when sarcasm is involved. Each tweet is typically analyzed by a fixed number of workers and majority voting. We here propose a crowdsourcing framework that instead uses a dynamic allocation of the number of workers. We explore two dynamic-allocation methods: (1) The number of workers queried to label a tweet is computed offline based on the predicted difficulty of discerning the sentiment of a particular tweet. (2) The number of crowd workers is determined online, during an iterative crowd sourcing process, based on inter-rater agreements between labels.We applied our approach to 1,000 twitter messages about the four U.S. presidential candidates Clinton, Cruz, Sanders, and Trump, collected during February 2016. We implemented the two proposed methods using decision trees that allocate more crowd efforts to tweets predicted to be sarcastic. We show that our framework outperforms the traditional static allocation scheme. It collects opinion labels from the crowd at a much lower cost while maintaining labeling accuracy.

Jianming Zhang, Shugao Ma, Mehrnoosh Sameki et al.

We study the problem of Salient Object Subitizing, i.e. predicting the existence and the number of salient objects in an image using holistic cues. This task is inspired by the ability of people to quickly and accurately identify the number of items within the subitizing range (1-4). To this end, we present a salient object subitizing image dataset of about 14K everyday images which are annotated using an online crowdsourcing marketplace. We show that using an end-to-end trained Convolutional Neural Network (CNN) model, we achieve prediction accuracy comparable to human performance in identifying images with zero or one salient object. For images with multiple salient objects, our model also provides significantly better than chance performance without requiring any localization process. Moreover, we propose a method to improve the training of the CNN subitizing model by leveraging synthetic images. In experiments, we demonstrate the accuracy and generalizability of our CNN subitizing model and its applications in salient object detection and image retrieval.

Mikhail Breslav, Tyson L. Hedrick, Stan Sclaroff et al.

Our work introduces a novel way to increase pose estimation accuracy by discovering parts from unannotated regions of training images. Discovered parts are used to generate more accurate appearance likelihoods for traditional part-based models like Pictorial Structures [13] and its derivatives. Our experiments on images of a hawkmoth in flight show that our proposed approach significantly improves over existing work [27] for this application, while also being more generally applicable. Our proposed approach localizes landmarks at least twice as accurately as a baseline based on a Mixture of Pictorial Structures (MPS) model. Our unique High-Resolution Moth Flight (HRMF) dataset is made publicly available with annotations.